Hi,
I am a student in the UK. As part of my GCSE Electronics major project
I am designing and making an automatic signalling system for a model
railway. I need to design a questionnaire to research what modellers
want in a signalling system. I have posted the short questionnaire at
http://uk.geocities.com/j_pilborough/qs.html . I would be very grateful
if some of you could fill it in!
Thanks in advance.
[8D]
I filled out the survey. Good luck on the project.
Andrew
J;
For a North American signal system, four aspects is a bare minimum. I would strongly recommend that you read the recent series in MR on CMRI
See: http://index.mrmag.com/tm.exe?opt=S&cmdtext=chubb&MAG=ANY&output=3&sort=2
Not discussed:
Needs Common rail (for detection)?
Detection of train in block with power turned off?
Sensing and signalling switch thrown against?
This is a UK system, however there are few UK based forums for modellers.
The choices for sensing are basically reed switches, or infrared. I know about using track circuits which can detect trains without using sensors, which I think is what you are referring to, but these would not be feasible with my current electronics knowledge. The system is also intended to be for straight runs only, not dealing with junctions. As this is a school project, I am unfortunately forced to spend large amouts opf time doing things like isomeric box designs and evaluations with brilliant presentation, rather than spending time designing the system. I am also limited to 20 hours practical work, including building a circuit box. I intend to post the design work and circuit designs on the interent when I am finished.
Thanks for your help. I now have about 15 replys.
JP.
By the way, a while ago there was an article in MR about computer signalling, where the system could be infinitly expanded by plugging each new module into the exisiting module on the end of the chain. Can anyone remember the URL for it?
JP
Hi j… pilborough
You are aware I take it that UK signaling is now four aspect?
I filled in your survey would not be buying as it requires magnets intead of resistors to be fitted to rolling stock.
However I would think for most layouts two or three aspect would be good enough.
As I dont know any one who has enough room for other than arrival and departure signals and if they are lucky maybe one set of intermediates
have you concidered interlocking and route indicators in your system.
Will it work just as well with semaphores the LBSCR. GNR. LT&SER did not have colour lights and if your system doesn’t cater for both you have a possable marketing problem to sort out.
That should give you enough to think about for your GCSE
hope this helps
For the benefit of my US friends the railways quoted are English ones so there isnt a second R
regards John
John, why would I want to fit resistors to rolling stock?
My design brief is to design a system for a large layout e.g. garden which would need several intermediate signals.
I am considering a modular system e.g. sensing module plugs into adjacent sensing modules and into signal driver which could be semaphore or colour light, 2, 3 or 4 aspect. The problem is I do not have the time or resources to develop alternative modules. The other problem with semaphores is that solenoids to operate them with are expensive.
I realised after I had uploaded the form that it was a bit silly to put a 1 apect option on it. However, this has given me another idea for a module - GWR searchlight signal!
Approximately 1/3rd of views of the form have come from this forum.
JP
there is an extreamly usefull website devoted to more advanced model railway electronics. only a small part of this site is for DCC and the circuits are excellant (i have built many and they have all so far worked.
http://home.cogeco.ca/~rpaisley4/CircuitIndex.html#index
the train detectors are particularly good.
good luck with your project (i did a similar thing for my degree) let me know how it goes.
Peter
The resistor provides a load and completes a circuit between the rails sufficient for a detector to pick up the voltage drop. A cut of cars thus equipped, or even a single car, does not have to be directly over an infrared detector for the signal logic to “know” the block is occupied and thus not clear a signal into the block. Reed switches may be sufficient for causing a signal to “drop” to all reds but are impractical for occupancy detection.
I understand that for the project you’re working on you probably don’t need to represent prototypical block occupancy, but it would be helpful if you decide to expand on the project at a later time.